PHYSICAL AI SECURITY & SAFETY
物理 AI 的安全
从硬件锚点开始
Security for physical AI,
anchored in hardware
为无人机与人形机器人
定义并交付硬件级零信任安全解决方案
We define and deliver hardware-level zero-trust security solutions for drones and humanoid robots
AI 进入物理世界
网络攻击可以直接造成物理伤害
Once AI acts in the physical world, a cyberattack can cause physical harm
"将网络安全视为任何能在物理世界中运作的系统的核心工程要求，而非最后附加的合规层"
"…treating cybersecurity as a core engineering requirement for any system that can act in the physical world, not as a compliance layer added at the end of development"
世界经济论坛，2026 年 10 月 9 日 World Economic Forum, 9 October 2026
这一要求正在写进法规
That requirement is now being written into regulation
欧盟《机械法规》(EU) 2023/1230 · 附件 III 第 1.1.9 条
EU Machinery Regulation (EU) 2023/1230 · Annex III, 1.1.9
机械的设计与制造须确保：连接其他设备或远程设备不会导致危险情形；与安全合规相关的硬件和软件须防范意外或蓄意的篡改。
Machinery must be designed so that connecting another device, including a remote one, cannot create a hazardous situation, and safety-critical hardware and software must be protected against accidental or intentional corruption.
2027 年 1 月 20 日起适用 · 条文转述 · EUR-Lex
Applies from 20 January 2027 · Paraphrased · EUR-Lex
欧盟《网络韧性法案》(EU) 2024/2847 · 附件 I 第一部分第 (1) 项
EU Cyber Resilience Act (EU) 2024/2847 · Annex I, Part I (1)
带数字元素的产品须在设计、开发和生产中，确保与风险相称的网络安全水平。
Products with digital elements must be designed, developed and produced to ensure a level of cybersecurity appropriate to the risks.
2027 年 12 月 11 日起适用 · 条文转述 · EUR-Lex
Applies from 11 December 2027 · Paraphrased · EUR-Lex
让信任可被验证
Make trust verifiable
零信任
不默认信任任何设备、固件或指令，每一次都验证
Zero trust
No device, firmware or command is trusted by default — each is verified, every time
源头安全
安全在设计时植入，不在出事后修补
Secure at the source
Built in at design time, not patched after an incident
硬件锚定
身份、密钥与启动链锚定在安全硬件中，使身份可信、固件防篡改、指令可验证，并可由第三方独立验证
Anchored in hardware
Identity, keys and the boot chain are anchored in secure hardware — trusted identity, tamper-resistant firmware and verifiable commands that third parties can check independently
承安智序 2026 年创办于上海，创始团队来自清华大学技术背景，专注物理 AI 的硬件级安全解决方案
Founded in Shanghai in 2026 by a team with a technical background from Tsinghua University, Cortenna Intelligence builds hardware-level security solutions for physical AI